PROJECT SUMMARY/ABSTRACT Alzheimer?s disease and related dementias are marked by early and significant impairments in decision- making, often with disastrous consequences for patients and their families. While recent research in decision neuroscience and neuroeconomics holds great promise for elucidating neural mechanisms underlying these impairments (and ultimately, for achieving a deeper understanding that can yield strategies for improving patients? decisions), in many cases it remains unclear how neuroscientific findings can be applied to the decision-making errors that patients exhibit in real life. This translational gap between cutting-edge research and clinical practice will be addressed in the proposed work by linking current decision neuroscience to clinical standards routinely applied in the assessment of patients? decision-making abilities. On a widely- accepted standard, impaired decisions in dementia and other cognitive disorders can result from failures to: (1) understand relevant features of a decision, (2) appreciate how this relates to one?s own situation, (3) rationally manipulate information to arrive at a decision, or (4) consistently evidence a choice. These criteria reflect clinicians? expertise with patients? impairments and embody mechanistic assumptions about how people make decisions and how decision-making is compromised in disease?assumptions that can now be investigated empirically using tools of decision neuroscience. The proposed research is organized around three specific aims that represent links between current neuroscience and accepted clinical standards: 1 (Appreciation): Investigate metacognitive impairments in different dementia syndromes in value-based choice. 2 (Reasoning): Elucidate mechanisms for applying relevant information to decision-making, which are disrupted in Alzheimer?s disease. 3 (Choice): Assess inconsistency in revealed preferences in different dementia syndromes. In this work we will draw from well-characterized cohorts of patients with Alzheimer?s disease, patients with frontotemporal dementia, and healthy older controls, all with linked clinical and neuroimaging data. Multiple cohorts will allow for specific associations between syndromes or targeted brain structures and abnormal decisions; otherwise, generic effects of illness or diminished cognition often cannot be excluded. Aims 1 & 3 will be pursued using computer-based behavioral testing with decision-making tasks, while Aim 2 will be pursued using task-based fMRI during a decision-making task. The innovative approach will enhance external validity (by linking results to the work of clinical experts about the decisional impairments that patients exhibit in daily life) and rigor (using comparisons across diseases to assure that findings reflect specific neural effects rather than other confounds). The proposed work is therefore poised to make a significant contribution by linking decision neuroscience with widely-accepted clinical standards to advance our understanding of mechanisms of clinical decision-making impairments, ultimately resulting in improved assessment tools and in targets for future interventions to prevent serious harms to patients.